ANNALEN DER PHYSIK 7. FOLGE * BAND 27, HEFT 1 * 1971 Quantum Mechanics as a Theory of Relativistic 6RowNian Motion By Yu.A.RYLOV Abstract It is shown that non-relativistic quantum mechanics can be treated as a kind of relati- vistic statistical theory which describes a random motion of a classical particle. The theory is relativistic in the sense that for the description of the particle behaviour the relativistic notion of the state is used. This is very important because a statistics is a state calculus and the result depends on the definition of the notion of state. Attempts by different authors [1-1411) to treat quantum mechanics from the position of classical theory did not lead to full success2). Though it may seem paradoxical, the reason there of, in our opinion, lies in the fact that the attempts to understand non-relativistic quantum mechanics were based on non-relati- vistic classical mechanics. In this paper we shall show that quantum mechanics is a variety of the re- lativistic theory of BRowNian motion3). The difference of our approach from approaches of others lies in the fact that we treat non-relativistic quantum mechanics from the point of view of relativistic classical mechanics. New principles are not needed for understanding the quantum mechanics of a single particle in this approach. In particular, such specific quantum mechanical principles as the uncertainty and the correspondence principle may be under- stood from the classical position. The clue to such an approach is the relativistic notion of system state. 1. The notion of state In non-relativistic physics the state of a physical system is defined4) as a set of quantities which are given at a certain moment of time and determine these quantities at any subsequent moment of time. For this purpose equations of motion are used. They describe the time evolution of the system state. The state and the equations of motion describing the time evolution of the state are two essential elements of any non-relativistic physical theory. l) A more comprehensive bibliography in [15]. 2, The paper [IS] is an exeption. The success here was reached, it seems, because what we call a relativistic notion of the state has been taken into account. 3, We call BRowNian motion any indeterministic motion of particles irrespective of the causes of its indeterminism. 4, Sometimes the state is defined as a set of independent variables. For our purpose the independence is unessential. 1 Ann. Plijsik. 7. Folgr, Bd. 27 2 Annalen der Physik $ 7. Folge * Band 27, Heft 1 v 1971 As follows from the definition, the state of the system is given at a certain time moment. But in relativistic theory simultaneity is relative. Which events are synchronous and which are not depends on the choice of a frame of reference. If, for example, one knows a state of a physical system in a frame of reference K one could describe the state in a frame of reference K’ moving relative to K only in the case when the equations of motion are known and can be solved. Thus, in the relativistic theory the state and the equations of motion are connected closely. Because there is no absolute simultaneity in the relativistic theory it seems more consistent to define the state of a system not for a given moment but over all space-time. In this case the conception of state will include the law of evolution of the physical system. The equations of motion are treated now as constraints imposed on the possible states. From all possible states not all states are realized but only those which satisfy certain equations. We shall call them the constraint equations. In re- ality they are the same equations of motion but now they do not describe the time evolution of the state but are restrictions which choose the physically allowable states from the virtual ones. In short, in the non-relativistic theory the unique division of the physical phenomena description into states and equations of motion corresponds to the unique division of space-time into space and time. In the relativistic theory where the division of space-time into space and time is conventional and not unique the division of the physical phenomena description into states and equa- tions of motion is not unique either. The physical system state defined over all space-time corresponds much better to the indivisible space-time. The manner of division of the description of a physical system into states and equations of motion is unimportant for the dynamics but is important for the statistics because statistics is the calculus of states. It is important for sta- tistics what is understood by “state ”. In general, a statistics that corresponds to a different division of the description of a physical system into states and equations of motion leads to different results. Let us describe the state of a single particle by giving its world-line over all the space time, i.e. by giving four functions 9%= q2(t),i = 0,1,2,3, where z is some parameter along the world-line. We need not give the momenta, pro- vided the mass m of the particle and its state are known. If the world-line is known the momenta are determined by the relations where c is light speed, q,k the metric tensor ;lc2 0 0 O/ 1,o 0 0-1 1 In equation (1.1)and henceforth the latin subscripts take the values 0,1,2,3 and greek ones take 1,2,3. As usual summation is made on like super- and sub- scripts. Yu. A. RYLOV: Quantum Mechanics as a Theory of Relativistic BROwNian Motion 3 2. The quantum ensemble Let us suppose that the state of a particle, i.e. its world-line, is a random quantity. Let us take for the sake of simplicity that the world-line cannot turn back in the time direction, i.e. that dqo/dt always keeps its sign. To describe the random states let us introduce the notion of state density. We consider at a point qa of space-time an infinitesimal area dX,. It is evident that the number dN of world-lines which cross the area dS, is proportional to the value of the area, that is dN = jkddSk where jk is a factor. The vector jk is proportional to the density of the world- lines in the vicinity of the point q. According to definition jk is the state density vector. This is connected with the fact that the object of statistics are one-di- mensional lines and not points as in non-relativistic statistics. According to the definition of jk the time component jorepresents the mean density of particles and the space components j’ represent the mean density of the particle flux. Thus the difference of principle between the non-relativistic and relativistic statistics consists in the fact that, for the former the state density is a scalar while for the later it may be, for example, a vector. Because the state density is a vector we shall be able to represent quantum mechanics as a theory of the relativistic &owxian motion. Later on we shall consider only the non-relati- vistic case, adopting from relativity only the relativistic notion of state density. Let us consider now a statistical ensemble of world-lines. Suppose the case is non-relativistic, i.e. the world-lines derivate but little from some constant direction in space-time. We choose this direction as the time axis, then cjo 9 lj”/. (2.1) The ensemble state is described by the state density. This means that the statre of the ensemble is considered as the state of some deterministic physical system. The state of the system is determined by the vector jk. The statistical ensemble is a deterministic system with the help of which one can describe non- deterministic ones. In the non-relativistic case the state density W is a function defined in the phase space. W is the ensemble state in the sense that beeing given at a moment of the time it can be uniquely determined for any following moment. There is another aspect. W is a non-negative quantity and witha suitable normalization WdQ can be interpreted as the probability to find the particle in the volume element dQ of the phase space. Together with the fact that W is a state this fact gives us a chance to speak about the random MARxovian process. In general, the two aspects of a statistical ensemble are independent, i. e. the state of the ensemble, being a state density, can not be a probability density. In the relativistic case it is important that the statistical ensemble described by the vector jk is a deterministic physical system. The fact that jkdS, can be interpreted as a probability to find a particle in the 3-volume dS, is valid only when the world-lines of particles do not zigzag in time. For relativistic particles when the generation of paires is possible such an interpretation is not suitable. We shall obtain the equations for the vector jk. In the simplest case the en- semble consists of strait world-lines which do not cross each other. It describes the motion of a gas of zero temperature. In such a gas the velocity of a single 1* 4 Annalen der Physik * 7. Folge * Band 27, Heft 1 * 1971 molecule coincides with the mean velocity of the gas stream. The chaotic mo- tion of the molecules and their diffusion are absent and there is no pressure at all.